Review and Measurements of Okto DAC8 8Ch DAC & Amp

[Kal Rubinson]

Not directly, but the JBL SDP-55 has Dante, so one should be able to get AES/EBU out from there, for about $150 for each 2 channel pair.

Good point.

 

[slags]

The DAC8 doesn't do Atmos decoding or DSP - it's a pure multichannel DAC. You'd need a software decoder to run on the computer - if you find one (other than Dolby's Mac Pro only Media Producer Suite) there will be a number of interested people here! This is almost certainly a licensing restriction not a technical one.

Right. So can something like VLC not do the job for a bluray? Reading some contrary stuff online on this

 

[somebodyelse]

Right. So can something like VLC not do the job for a bluray? Reading some contrary stuff online on this

Unless I've missed something (quite possible!) VLC can do passthrough of Atmos via HDMI to an AVR, but can't decode it itself. It can decode some of the older formats though (AC3, DTS, others?) so may fall back to these if passthrough isn't available. If you know otherwise (new codec that hasn't been documented in the wiki for example) I'd love to hear about it.

 

[Snoozer]

Broken in what way? Mine should arrive soon, and I plan to connect it to a Linux PC.

I am sorry, just missed your post.

By broken I mean the firmware does not work in Linux (any Linux). There seems to be a problem within the XMOS firmware (it is the USB receiver sort to speak). If you try to use it in Windows, it will work with some restrictions to be taken into account.

I am still waiting for news on a revised new firmware version to solve those problems. Unfortunately for me the my unit is in store until there is a solution.

 

[DWPress]

Well, I just ordered mine today after thinking about it hard for over a year. I know it'll take some time to get here but I can wait a bit longer and, with luck, this will be the end of the journey for me as far as DACs go in my active XO system.

Look for some used Motu gear and a miniDSP 4x10 to be showing up in the for sale section soon before I list it on fleaBay.

 

[DWPress]

Anyone using this with a Mac OS system? Any issues I should be aware of? (I'm still on Mojave)

edit:
Also - What player software are you using and any issues with any of them and non-home theater multichannel use? I've set up an active XO system for my 3-way mains and 3 subs in both Roon and Jriver in the past (now actually) using the Motu or MiniDSP 4x10 but getting routing to work properly on the Motu wasn't fun to figure out.

I've also been toying with getting things to work system wide on the Mac using apps from Rogue Amoeba (SoundSource & Audio Hijack) or Pedalboard2 and do the XO and convolution via AU or VST plugins, host software needed anyway if I want to try the new Dirac. This way any app on my computer can have the benefit of the DSP/DRC including BT coming in via Airfoil.

 

[Honken]

I am sorry, just missed your post.

By broken I mean the firmware does not work in Linux (any Linux). There seems to be a problem within the XMOS firmware (it is the USB receiver sort to speak). If you try to use it in Windows, it will work with some restrictions to be taken into account.

I am still waiting for news on a revised new firmware version to solve those problems. Unfortunately for me the my unit is in store until there is a solution.

No worries.

I emailed Pavel about this and they are aware of the problem and have a fix in the pipe. It'd be nice if they had some kind of public changelog for their FW and offered tooling to flash the devices ourselves.

 

[somebodyelse]

I emailed Pavel about this and they are aware of the problem and have a fix in the pipe. It'd be nice if they had some kind of public changelog for their FW and offered tooling to flash the devices ourselves.

Agreed about the changelog. If they're sticking to the standard XMOS update code (which pretty much everyone seems to) then a linux cli updater is trivial based on the XMOS example Mac updater - it's already been used for the Tone Board, at least one Topping device and possibly a Soncoz. We can provide Pavel with the instructions if necessary - given the price I expect people to be much happier with a vendor-supported updater!

 

[Feyire]

It'd be nice if they had some kind of public changelog for their FW and offered tooling to flash the devices ourselves.

@Okto Research make it happen

 

[fabriceo]

Hi Guys,
I m a friend of Pavel and I m involved in the XMOS implementation on dac8pro and dac8stereo.
So far this was a fascinating project as the implementation made for Okto research is quite special for handling the 4 spdif signals in the 3 different modes...
Because I am the proud owner of a unit and because I was dreaming for an active crossover embedded in the device for ages, I made it
and it s running now at home for 6 months on a pair of 2 way speakers (12"+2"compression).
I ve chosen a "delayed subtractive bessel 6th" crossover and the result is exceptional with some EQ;

by this mail I thought I could give to some of you the opportunity to implement this solution in their dac.
its a free and diy approach not supported directly by Okto research, but by me as a diyer
No commercial product are planed with this feature as far as I know, due to the related complexity to implement it.

the solution is based on using the "AVDSP" framework (audio-virtual-DSP) which has been developed as a first step, to be platform agnostic solution (a friend use it on linux in alsa) and then a special bridge was made so that it is optimized and perfectly integrated to the XMOS audio task and running in multiple thread on the core 0 (core 1 is for usb).

the xmos 200 is a nice boy but still it has limitations. all depend on the number of tasks you can put in parallel for your DSP program and then depends on the number of SPDIF/AES channel you use. the other factor is the maximum speed (48/96/192k). I went for maximum 96k and this is what I d recommend if you go with it. no dsd capability in this dsp mode.
in a nutshell, each dsp task available as an xmos-core can handle 8 biquads and 1 delay and a gain at 192k for 2 channels, or twice at 96k.
treatment can be done from usb host to dac, or from spdif/aes to usb-host or to dac if you need that.

To provide adequate support, I d like to offer 2 candidates the possibility to make it and if successful we'll go forward with more participants; you are already familiar with active crossover and you can compile a c program in a command line?
if you are interested, you can post "briefly" your context here, but to avoid jeopardizing the thread too much, just drop me a PM!

cheers
fabriceo

 

[dualazmak]

As I am not an expert of C compiler on LINUX OS, I will not join the interesting and exciting development program you kindly offering.

I am very much looking forward to, however, the near-future "output" of your project (compiled exe DIY modules with nice GUI!) to be installed both in LINUX OS and Windows OS. Until then, I will continue to use the software crossover EKIO (192 kHz 24 bit) on 64 bit Windows 10 with my DAC8PRO (FW ver.1.32) through USB routing. If needed, I will be happy to be a beta tester for your possible new solutions on Windows in my multichannel multi-amplifier project. Good luck for your exciting developments...

 

[fabriceo]

Hi dualazmak, I saw your different post on this crossover with EKIO, do you mind sending me by PM the screenshot of your configuration as defined in EKIO (for each speaker) and I ll see what I can do or not. let me know if you use multiple AES/SPDIF or not (to free some tasks).

For those who want to know how this work, the crossover I used is described with some macros in the file available on my GitHub here:
https://github.com/fabriceo/AVDSP/blob/master/module_avdsp/dspprogs/oktodac.c
it stands between lines 399 and 530. The workload is spread cross 3 tasks/cores. okay, this is not the easiest one
output 1&2 are stereo passthrough for my headset, 3,4,5,6 are the 2x2 output, 7 for subwoofer and 8 for a center channel.
This c file has to be compiled with gcc as a dynamic library, then a command line utility will use this library and some parameters (fx, gain) to generate a static "bin" file containing all the dsp opcodes and biquad coefficients for all the supported frequencies, and then this file can be uploaded in the flash memory of the xmos, with another command line tool. that's it all the tests and tuning were done with REW using digital loopback before turning on the amps of course.

 

[dualazmak]

Hi dualazmak, I saw your different post on this crossover with EKIO, do you mind sending me by PM the screenshot of your configuration as defined in EKIO (for each speaker) and I ll see what I can do or not. let me know if you use multiple AES/SPDIF or not (to free some tasks).
........

Hello fabriceo,

Sure sir!

Although I am sharing my latest EKIO configuration at my post here, I will send you some more screen shots tonight through PM system.

As shown there, I am using just the AES/EBU digital out of DAC8PRO (CH1+CH2) into another DAC, ONKYO DAC-1000, for RCA unbalanced input into my active sub-woofers YAMAHA YST-SW1000 (L and R).

 

[DWPress]

treatment can be done from usb host to dac, or from spdif/aes to usb-host or to dac if you need that.

I've briefly looked at the code. I'm on OSX and if you figure out the CoreAudio port I'd be very interested to trial. This would make XO and (possibly?) some dsp active system wide correct? There are other VST plugins out there that will do XO, how does this differ?

 

[Zooqu1ko]

I've briefly looked at the code. I'm on OSX and if you figure out the CoreAudio port I'd be very interested to trial. This would make XO and (possibly?) some dsp active system wide correct? There are other VST plugins out there that will do XO, how does this differ?

If I understand this correctly, this will run the crossover entirely in the DAC 8, with no OS involvement at all. The DAC8 would then look essentially the same to the source (USB or AES/EBU), but accept fewer input channels (e.g. just 2), but cross them over at the desired frequencies and send the split signal to the various analog outputs (or upstream via USB). No new driver needed.

 

[Vasr]

Hi dualazmak, I saw your different post on this crossover with EKIO, do you mind sending me by PM the screenshot of your configuration as defined in EKIO (for each speaker) and I ll see what I can do or not. let me know if you use multiple AES/SPDIF or not (to free some tasks).

For those who want to know how this work, the crossover I used is described with some macros in the file available on my GitHub here:
https://github.com/fabriceo/AVDSP/blob/master/module_avdsp/dspprogs/oktodac.c
it stands between lines 399 and 530. The workload is spread cross 3 tasks/cores. okay, this is not the easiest one
output 1&2 are stereo passthrough for my headset, 3,4,5,6 are the 2x2 output, 7 for subwoofer and 8 for a center channel.
This c file has to be compiled with gcc as a dynamic library, then a command line utility will use this library and some parameters (fx, gain) to generate a static "bin" file containing all the dsp opcodes and biquad coefficients for all the supported frequencies, and then this file can be uploaded in the flash memory of the xmos, with another command line tool. that's it all the tests and tuning were done with REW using digital loopback before turning on the amps of course.

No doubt this would satisfy some use cases but if you really want to make this useful to a larger audience I recommend making it possible to use this as a 7.1 ch (or 5.2) with crossover.

With this capability and the volume balance, the Okto 8 Pro becomes a good pre/pro for 7.1 ch use via USB input and people can add whatever room correction they want upstream.

The requirements to be able to do this is to be able to specify a crossover point and slope for each channel, HF or LF pass for each channel and the routing of the split to the channel you want and the ability to add the split content coming to the same channel.

So, for a 7.1, you would have high pass for 7 channels and low pass for 1 with user selected slopes and crossover point and the LF content from each of the 7 channels added to the content coming into the 8th channel. For extra points, you can provide a switch to add 10db analog boost to the 8th channel for HT use.

Not asking you to do it necessarily. Don't know if this is technically feasible and how you would do this without a user interface, just stating what the requirements would be if this were to be useful to a larger audience - and to cater to a space (multi-channel audio/video) that is very under-served by good DACs at the moment.

 

[haraldo]

If I understand this correctly, this will run the crossover entirely in the DAC 8 ......

did I miss something here?
the dac 8 pro has no x-over functions, according to specs

 

[Zooqu1ko]

did I miss something here?
the dac 8 pro has no x-over functions, according to specs

It's inofficial, see fabriceo's post #1,771, just 7 posts before your own.

 

[haraldo]

It's inofficial, see fabriceo's post #1,771, just 7 posts before your own.

Thx, wow

 

[phoenixdogfan]

No doubt this would satisfy some use cases but if you really want to make this useful to a larger audience I recommend making it possible to use this as a 7.1 ch (or 5.2) with crossover.

With this capability and the volume balance, the Okto 8 Pro becomes a good pre/pro for 7.1 ch use via USB input and people can add whatever room correction they want upstream.

The requirements to be able to do this is to be able to specify a crossover point and slope for each channel, HF or LF pass for each channel and the routing of the split to the channel you want and the ability to add the split content coming to the same channel.

So, for a 7.1, you would have high pass for 7 channels and low pass for 1 with user selected slopes and crossover point and the LF content from each of the 7 channels added to the content coming into the 8th channel. For extra points, you can provide a switch to add 10db analog boost to the 8th channel for HT use.

Not asking you to do it necessarily. Don't know if this is technically feasible and how you would do this without a user interface, just stating what the requirements would be if this were to be useful to a larger audience - and to cater to a space (multi-channel audio/video) that is very under-served by good DACs at the moment.

Would also be great if it had an 8 band peq for each output channel, and was as simple as miniDSP to implement.